That is a good point, but I think the comparison is accurate. I've run the tests hundreds of times with the performance difference always being the same.
-Todd On Mon, Oct 06, 2003 at 12:43:08PM -0400, Geoffrey Young wrote: > > >My application seemed slower with Resistry! Using curl to get more > >accurate statistics I found Registry to be almost 2-3x slower than > >PerlRun. > > > >### BEGIN sample stats ### > >[EMAIL PROTECTED] root]# date +%M.%S.%N; curl -b cookies.txt -c cookies.txt > >http://usvwoaahs52:8080/serverdb/cgi-bin/devlist.pl > /dev/null; date > >+%M.%S.%N; > >29.41.134877000 > > % Total % Received % Xferd Average Speed Time > > Curr. > > Dload Upload Total Current Left > > Speed > >100 421 0 421 0 0 295 0 --:--:-- 0:00:01 --:--:-- > >1042 > >29.42.581020000 > > you can't possibly get an accurate comparison from a test that takes only a > few seconds seconds to execute. > > look into using ab (apache bench, distributed with apache) or Benchmark.pm > for your benchmarks instead, and make sure that you make a significant > number of requests. > > another good resource for this kind of thing is here > > http://www.chamas.com/bench/ > > HTH > > --Geoff